Installation

Installation

The preferred installation mechanism for msmbuilder is with conda.

$ conda install -c https://conda.binstar.org/omnia msmbuilder

If you don’t have conda, or are new to scientific python, we recommend that you download the Anaconda scientific python distribution.

Dependencies

MSMBuilder is a python package that heavily leans on other components of the scientific python ecosystem. To use msmbuilder, the following libraries and software will need to be installed.

Linux, Mac OS X, or Windows
We develop mainly on 64-bit Linux and Mac machines. Windows _is_ supported, but may be a little rough around the edges.
Python >= 2.6
The development package (python-dev or python-devel on most Linux distributions) is recommended.
NumPy >= 1.6.0
Numpy is the base package for numerical computing in python.
SciPy >= 0.11.0
We use scipy for sparse matrix, numerical linear algebra and optimization.
scikit-learn >= 0.14.0
Many of the models in msmbuilder build off base classes in scikit-learn.
MDTraj >= 1.2.0
MDTraj is a library for handing molecular dynamics trajectories.
Pandas >= 0.9.0
Pandas is pretty cool
cython >= 0.18.0
This is needed to compile the package.
cvxopt
Only one module in MSMBuilder uses cvxopt. TODO: make cvxopt optional.

Optional packages:

nose
Recommended, to run the test-suite.
Git
To download the source code

Frequently Asked Questions

I get a wierd error during compilation with pip install or python setup.py install. What’s wrong?

We’ve received a couple reports of incorrect code generation on earlier versions of cython with python3.4. Try upgrading to the latest verion of cython, and reinstalling. If that doesn’t work, open an issue on the github issue tracker.

Do I need Anaconda python? Can’t I use the python that comes with my operating like /usr/bin/python?

You can have multiple python installations on your computer which do not interact with one another at all. The system python interpreter is used by your operating system for some of its own programs, but is not the best choice for data analysis or science.

We strongly recommend that you install Anaconda or Miniconda python distribution and that you have the conda package manager available.

If you’re interested in some of the details about packaging and scientific python, see this blog post by Travis Oliphant.

How do I get (quickly) started on Windows?

As mentioned above, our Windows support is rough around the edges. However, as of MSMBuilder 3.0, we have conda packages for (only!) win32 + py34. Therefore, one path to get up and running is:

  1. Download and Install Continuum Anaconda 32 bit py34 (Note: we don’t currently have packages for 64 bit or py27!)
  2. Run Anaconda Command Prompt
  3. cd Scripts
  4. conda install -c https://conda.binstar.org/omnia msmbuilder
  5. Close the command prompt
  6. Run Anaconda ipython3.4 qtconsole from the start menu–you should be ready to roll!
Versions